More usage of TclUtfToUCS4(), so we can use the whole Unicode range better in TCL_UTF_MAX>3 builds.

4 files changed, 36 insertions, 36 deletions

diff --git a/generic/tclCmdIL.c b/generic/tclCmdIL.c
index 7e685bd..3ec1c09 100644
--- a/generic/tclCmdIL.c
+++ b/generic/tclCmdIL.c
@@ -4370,7 +4370,7 @@ static int
 DictionaryCompare(
     const char *left, const char *right)	/* The strings to compare. */
 {
-    Tcl_UniChar uniLeft = 0, uniRight = 0, uniLeftLower, uniRightLower;
+    int uniLeft = 0, uniRight = 0, uniLeftLower, uniRightLower;
     int diff, zeros;
     int secondaryDiff = 0;
 
@@ -4439,8 +4439,8 @@ DictionaryCompare(
 	 */
 
 	if ((*left != '\0') && (*right != '\0')) {
-	    left += TclUtfToUniChar(left, &uniLeft);
-	    right += TclUtfToUniChar(right, &uniRight);
+	    left += TclUtfToUCS4(left, &uniLeft);
+	    right += TclUtfToUCS4(right, &uniRight);
 
 	    /*
 	     * Convert both chars to lower for the comparison, because
@@ -4449,8 +4449,8 @@ DictionaryCompare(
 	     * other interesting punctuations occur).
 	     */
 
-	    uniLeftLower = Tcl_UniCharToLower(uniLeft);
-	    uniRightLower = Tcl_UniCharToLower(uniRight);
+	    uniLeftLower = TclUCS4ToLower(uniLeft);
+	    uniRightLower = TclUCS4ToLower(uniRight);
 	} else {
 	    diff = UCHAR(*left) - UCHAR(*right);
 	    break;
diff --git a/generic/tclInt.h b/generic/tclInt.h
index 8983659..7fc06c8 100644
--- a/generic/tclInt.h
+++ b/generic/tclInt.h
@@ -3185,6 +3185,7 @@ MODULE_SCOPE int	TclTrimRight(const char *bytes, int numBytes,
 MODULE_SCOPE int	TclUtfCasecmp(const char *cs, const char *ct);
 MODULE_SCOPE int	TclUtfToUCS4(const char *, int *);
 MODULE_SCOPE int	TclUCS4ToUtf(int, char *);
+MODULE_SCOPE int	TclUCS4ToLower(int ch);
 #if TCL_UTF_MAX == 4
     MODULE_SCOPE int	TclGetUCS4(Tcl_Obj *, int);
 #else
diff --git a/generic/tclUtf.c b/generic/tclUtf.c
index 5e9b7a1..9792071 100644
--- a/generic/tclUtf.c
+++ b/generic/tclUtf.c
@@ -88,7 +88,6 @@ static const unsigned char totalBytes[256] = {
 static int		UtfCount(int ch);
 static int		Invalid(unsigned char *src);
 static int		UCS4ToUpper(int ch);
-static int		UCS4ToLower(int ch);
 static int		UCS4ToTitle(int ch);
 
 /*
@@ -1078,7 +1077,7 @@ Tcl_UtfToLower(
     src = dst = str;
     while (*src) {
 	len = TclUtfToUCS4(src, &ch);
-	lowChar = UCS4ToLower(ch);
+	lowChar = TclUCS4ToLower(ch);
 
 	/*
 	 * To keep badly formed Utf strings from getting inflated by the
@@ -1149,7 +1148,7 @@ Tcl_UtfToTitle(
 	lowChar = ch;
 	/* Special exception for Georgian Asomtavruli chars, no titlecase. */
 	if ((unsigned)(lowChar - 0x1C90) >= 0x30) {
-	    lowChar = UCS4ToLower(lowChar);
+	    lowChar = TclUCS4ToLower(lowChar);
 	}
 
 	if (len < UtfCount(lowChar) || ((lowChar & ~0x7FF) == 0xD800)) {
@@ -1254,11 +1253,11 @@ Tcl_UtfNcmp(
 	if (ch1 != ch2) {
 #if TCL_UTF_MAX == 4
 	    /* Surrogates always report higher than non-surrogates */
-	    if (((ch1 & 0xFC00) == 0xD800)) {
-	    if ((ch2 & 0xFC00) != 0xD800) {
+	    if (((ch1 & ~0x3FF) == 0xD800)) {
+	    if ((ch2 & ~0x3FF) != 0xD800) {
 		return ch1;
 	    }
-	    } else if ((ch2 & 0xFC00) == 0xD800) {
+	    } else if ((ch2 & ~0x3FF) == 0xD800) {
 		return -ch2;
 	    }
 #endif
@@ -1427,8 +1426,8 @@ Tcl_UniCharToUpper(
  *----------------------------------------------------------------------
  */
 
-static int
-UCS4ToLower(
+int
+TclUCS4ToLower(
     int ch)			/* Unicode character to convert. */
 {
     if (!UNICODE_OUT_OF_RANGE(ch)) {
@@ -1447,7 +1446,7 @@ Tcl_UniChar
 Tcl_UniCharToLower(
     int ch)			/* Unicode character to convert. */
 {
-    return (Tcl_UniChar) UCS4ToLower(ch);
+    return (Tcl_UniChar) TclUCS4ToLower(ch);
 }
 
 /*
diff --git a/generic/tclUtil.c b/generic/tclUtil.c
index 13b0d55..b4a07bb 100644
--- a/generic/tclUtil.c
+++ b/generic/tclUtil.c
@@ -2155,7 +2155,7 @@ Tcl_StringCaseMatch(
     int nocase)			/* 0 for case sensitive, 1 for insensitive */
 {
     int p, charLen;
-    Tcl_UniChar ch1 = 0, ch2 = 0;
+    int ch1 = 0, ch2 = 0;
 
     while (1) {
 	p = *pattern;
@@ -2196,12 +2196,12 @@ Tcl_StringCaseMatch(
 	     */
 
 	    if (UCHAR(*pattern) < 0x80) {
-		ch2 = (Tcl_UniChar)
+		ch2 = (int)
 			(nocase ? tolower(UCHAR(*pattern)) : UCHAR(*pattern));
 	    } else {
-		Tcl_UtfToUniChar(pattern, &ch2);
+		TclUtfToUCS4(pattern, &ch2);
 		if (nocase) {
-		    ch2 = Tcl_UniCharToLower(ch2);
+		    ch2 = TclUCS4ToLower(ch2);
 		}
 	    }
 
@@ -2215,8 +2215,8 @@ Tcl_StringCaseMatch(
 		if ((p != '[') && (p != '?') && (p != '\\')) {
 		    if (nocase) {
 			while (*str) {
-			    charLen = TclUtfToUniChar(str, &ch1);
-			    if (ch2==ch1 || ch2==Tcl_UniCharToLower(ch1)) {
+			    charLen = TclUtfToUCS4(str, &ch1);
+			    if (ch2==ch1 || ch2==TclUCS4ToLower(ch1)) {
 				break;
 			    }
 			    str += charLen;
@@ -2229,7 +2229,7 @@ Tcl_StringCaseMatch(
 			 */
 
 			while (*str) {
-			    charLen = TclUtfToUniChar(str, &ch1);
+			    charLen = TclUtfToUCS4(str, &ch1);
 			    if (ch2 == ch1) {
 				break;
 			    }
@@ -2243,7 +2243,7 @@ Tcl_StringCaseMatch(
 		if (*str == '\0') {
 		    return 0;
 		}
-		str += TclUtfToUniChar(str, &ch1);
+		str += TclUtfToUCS4(str, &ch1);
 	    }
 	}
 
@@ -2254,7 +2254,7 @@ Tcl_StringCaseMatch(
 
 	if (p == '?') {
 	    pattern++;
-	    str += TclUtfToUniChar(str, &ch1);
+	    str += TclUtfToUCS4(str, &ch1);
 	    continue;
 	}
 
@@ -2265,17 +2265,17 @@ Tcl_StringCaseMatch(
 	 */
 
 	if (p == '[') {
-	    Tcl_UniChar startChar = 0, endChar = 0;
+	    int startChar = 0, endChar = 0;
 
 	    pattern++;
 	    if (UCHAR(*str) < 0x80) {
-		ch1 = (Tcl_UniChar)
+		ch1 = (int)
 			(nocase ? tolower(UCHAR(*str)) : UCHAR(*str));
 		str++;
 	    } else {
-		str += Tcl_UtfToUniChar(str, &ch1);
+		str += TclUtfToUCS4(str, &ch1);
 		if (nocase) {
-		    ch1 = Tcl_UniCharToLower(ch1);
+		    ch1 = TclUCS4ToLower(ch1);
 		}
 	    }
 	    while (1) {
@@ -2283,13 +2283,13 @@ Tcl_StringCaseMatch(
 		    return 0;
 		}
 		if (UCHAR(*pattern) < 0x80) {
-		    startChar = (Tcl_UniChar) (nocase
+		    startChar = (int) (nocase
 			    ? tolower(UCHAR(*pattern)) : UCHAR(*pattern));
 		    pattern++;
 		} else {
-		    pattern += Tcl_UtfToUniChar(pattern, &startChar);
+		    pattern += TclUtfToUCS4(pattern, &startChar);
 		    if (nocase) {
-			startChar = Tcl_UniCharToLower(startChar);
+			startChar = TclUCS4ToLower(startChar);
 		    }
 		}
 		if (*pattern == '-') {
@@ -2298,13 +2298,13 @@ Tcl_StringCaseMatch(
 			return 0;
 		    }
 		    if (UCHAR(*pattern) < 0x80) {
-			endChar = (Tcl_UniChar) (nocase
+			endChar = (int) (nocase
 				? tolower(UCHAR(*pattern)) : UCHAR(*pattern));
 			pattern++;
 		    } else {
-			pattern += Tcl_UtfToUniChar(pattern, &endChar);
+			pattern += TclUtfToUCS4(pattern, &endChar);
 			if (nocase) {
-			    endChar = Tcl_UniCharToLower(endChar);
+			    endChar = TclUCS4ToLower(endChar);
 			}
 		    }
 		    if (((startChar <= ch1) && (ch1 <= endChar))
@@ -2350,10 +2350,10 @@ Tcl_StringCaseMatch(
 	 * each string match.
 	 */
 
-	str += TclUtfToUniChar(str, &ch1);
-	pattern += TclUtfToUniChar(pattern, &ch2);
+	str += TclUtfToUCS4(str, &ch1);
+	pattern += TclUtfToUCS4(pattern, &ch2);
 	if (nocase) {
-	    if (Tcl_UniCharToLower(ch1) != Tcl_UniCharToLower(ch2)) {
+	    if (TclUCS4ToLower(ch1) != TclUCS4ToLower(ch2)) {
 		return 0;
 	    }
 	} else if (ch1 != ch2) {